Return to BSD News archive
Xref: sserve comp.org.eff.talk:9190 comp.unix.bsd:5786 comp.os.mach:2221 misc.int-property:544 comp.sys.mac.advocacy:714 Newsgroups: comp.org.eff.talk,comp.unix.bsd,comp.os.mach,misc.int-property,alt.suit.att-bsdi,comp.sys.mac.advocacy Path: sserve!manuel!munnari.oz.au!sgiblab!sdd.hp.com!decwrl!concert!samba!SunSite.unc.edu!jem From: jem@SunSite.unc.edu (Jonathan Magid) Subject: Re: Letter asking for help with Apple from the US VP Message-ID: <1992Sep29.070054.12001@samba.oit.unc.edu> Sender: usenet@samba.oit.unc.edu Nntp-Posting-Host: sunsite.unc.edu Organization: University of North Carolina, Chapel Hill References: <1992Sep26.163059.24740@rwwa.COM> <29581.Sep2900.20.3792@virtualnews.nyu.edu> <v_+pvad.mcgregor@netcom.com> Date: Tue, 29 Sep 1992 07:00:54 GMT Lines: 46 In article <v_+pvad.mcgregor@netcom.com> mcgregor@netcom.com (Scott Mcgregor) writes: >In article <29581.Sep2900.20.3792@virtualnews.nyu.edu> brnstnd@nyu.edu (D. J. Bernstein) writes: > >>In 1984, Miller and Wegman at IBM invented a compression algorithm often >>called MW2. At the same time, independently, Welch at Unisys invented a >>compression algorithm often called LZW. Both IBM and Unisys filed for, >>and received, patents. Surprise: MW2 and LZW are the same algorithm. >>Explain how innovation was encouraged here. > [much stuff deleted] > >This is also true in the Storer/Bernstein case. What may be different >in that case is that we know that IBM and Unisys were motivated by the >possibile advantages of developing products in a patentable area >(after all, they also paid the costs to secure patents. [even more stuff deleted] Just a word on this point specifically. Although I have never been an employee of IBM, I know that it is a solid IBM policy to patent any possible thing that comes out of its research. It gives large incentive to its employees to develop patentable things (in the form of "patent points") which it patents rather indescriminately. Employees get bonus "points" for having a patent used in a IBM product. There several things that should be concluded from this: 1) IBM doesn't do research in an area to specifically claim patents; it looks for patents from all areas of corporate research. 2) IBM's policy on many of these patents (esp. software/algorithm patents) is that they are for the purpose of cross-licencing with other people that have layed claim to useful algorithms, etc. 3) IBM's corporate incentive to use patented processes and inventions in IBM products occasionally (at least) has negative consequences on IBM products. I know of a couple of anecdotes on products that either failed or performed badly because methods owned by IBM rather than the canonical method was used. jem. -- Jonathan Magid jem@sunsite.unc.edu SunSite Administrator Virtual pizza Delivery (tm)::faxed in 30 cycles or less or you get it ----------------------------------------FREE!!!-----------------------